Sharma et al. BMC Developmental Biology 2014, 14:34 http://www.biomedcentral.com/1471-213X/14/34

RESEARCH ARTICLE Open Access Changes in expression of Class 3 and their receptors during development of the rat retina and superior colliculus Anil Sharma1*, Chrisna J LeVaillant1, Giles W Plant1,2 and Alan R Harvey1

Abstract Background: Members of the 3 family (Sema3s) influence the development of the , and some are implicated in regulating aspects of visual system development. However, we lack information about the timing of expression of the Sema3s with respect to different developmental epochs in the mammalian visual system. In this time-course study in the rat, we document for the first time changes in the expression of RNAs for the majority of Class 3 Semaphorins (Sema3s) and their receptor components during the development of the rat retina and superior colliculus (SC). Results: During retinal development, transcript levels changed for all of the Sema3s examined, as well as Nrp2, Plxna2, Plxna3, and Plxna4a. In the SC there were also changes in transcript levels for all Sema3s tested, as well as Nrp1, Nrp2, Plxna1, Plxna2, Plxna3, and Plxna4a. These changes correlate with well-established epochs, and our data suggest that the Sema3s could influence retinal ganglion cell (RGC) apoptosis, patterning and connectivity in the maturing retina and SC, and perhaps guidance of RGC and cortical in the SC. Functionally we found that SEMA3A, SEMA3C, SEMA3E, and SEMA3F collapsed purified postnatal day 1 RGC growth cones in vitro. Significantly this is a developmental stage when RGCs are growing into and within the SC and are exposed to Sema3 ligands. Conclusion: These new data describing the overall temporal regulation of Sema3 expression in the rat retina and SC provide a platform for further work characterising the functional impact of these proteins on the development and maturation of mammalian visual pathways. Keywords: Retinal ganglion cells, Collapse assay, qPCR, , , Cell adhesion molecules

Background of visual pathways in mammals. It is likely then that other The mammalian visual system is complex, and the de- molecules are also involved during the development and velopmental events that give rise to this highly organised maturation of the mammalian visual system, and in this system are similarly complex and exquisitely ordered. context recent attention has turned to the Semaphorins. The timing of the major developmental events of the Semaphorins (Semas) have been implicated in neural mammalian visual system has been well characterised, and vascular aspects of visual system development in and many of the molecular cues controlling these critical various species including frog [8,9], zebrafish [10-14], events have been elucidated (for example Ephs/Ephrins, goldfish [15], and chicken [16-21]. In mouse the mem- Slits/Robo, Netrin/DCC, and various neurotrophic factors brane bound Class 5 and Class 6 Semaphorins are in- [1-7]). However despite our increased knowledge about volved in lamination of the retina [22-24], and guidance these molecular cues, they are not sufficient to explain of retinal ganglion cell (RGC) axons [25] via contact- completely the complexity and timing of the development mediated interactions. However many previously discov- ered molecular cues in the mammalian visual system are * Correspondence: [email protected] diffusible, and the only vertebrate secreted members of 1School of Anatomy, Physiology and Human Biology, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia the Semaphorins are the Class 3 Semaphorins (Sema3s) Full list of author information is available at the end of the article [26] which are known to be expressed in the developing

© 2014 Sharma et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Sharma et al. BMC Developmental Biology 2014, 14:34 Page 2 of 14 http://www.biomedcentral.com/1471-213X/14/34

rat retina [27]. It is possible then that the Sema3s also Changes in Class 3 Semaphorins and their receptor influence the complexity and timing of the developing components during rat retinal development mammalian visual system. Expression levels in the retina could be separated into three The Sema3s consist of Sema3a through Sema3g qualitative groups: relatively high expression of Sema3f and [16,26,28-36] and their main receptors are the Neuropilins Plxna2; moderate expression of Nrp1 and Plxna1;andrela- which form multimeric receptor complexes with Plexins tively low expression of the rest (Figures 2, and 3). There and cell adhesion molecules [37]. Sema3s were initially dis- were statistically significant changes in the level of expres- covered as guidance molecules [30,38], but are now sion of all Sema3 RNAs in the retina, while of the receptors known to also mediate apoptosis, cell migration, immune only Nrp2, Plxna2, Plxna3,andPlxna4a showed statisti- response, organogenesis, tumour suppression and promo- cally significant changes (Figures 2, and 3; Additional file 1: tion, and vasculature development [39-44]. Sema3s affect at Table S1). least some aspects of the development of the visual system Relative to other time points, Sema3a transcript ex- of rodents [45,46], from where much of our knowledge of pression levels were significantly increased at P14 and in molecular and activity-driven influences on mammalian the adult. Similar to Sema3a, Sema3b transcript expres- visual system development has come [6,47]. sion was relatively stable during retinal development In summary, while we understand much about the until P14, at which time there was increased expression timing of developmental events in the mammalian visual that was maintained into adulthood. Sema3c RNA ex- system, investigation of a potential role for the Sema3s pression was also relatively steady through to P0, in- in these events would benefit from knowledge about the creasing significantly through to P21, and remaining at timing of expression of Sema3s and their receptors dur- that level into the adult. Sema3e RNA levels appeared to ing visual system maturation. We sought to address this increase gradually with retinal maturation and were sig- gap in our knowledge by building on a previous study nificantly higher than E16-P7 levels at P21 and in adult from our laboratory [27], focusing on Sema3s known to rats. Sema3f transcription was temporarily greater at P0 be expressed in RGCs and extending the work to include and then increased again at P21 and beyond. Nrp2 RNA a greater number of receptors as well as analysis of a expression gradually increased throughout retina matur- major central target for those RGCs, the superior collicu- ation, levels at P7 and beyond being significantly greater lus (SC). To that end, we quantified transcript expression than in embryos. Plxna2 and Plxna3 showed nearly in the rat retina and SC over a range of embryonic (E) and identical patterns of altered transcript expression during postnatal (P) developmental time points (E16, E19, birth – development, both peaking significantly at the time of P0, P7, P14, P21 and adult) which were chosen to corres- birth (P0). Plxna4a RNA levels peaked later at P7, but pond to previously established developmental epochs such also again in the adult. as the timing of cell birth, migration, naturally occurring Many of the significant peaks in transcript expression cell death, axonal and dendritic growth, and synaptogene- occurred after the main developmental epochs. However, sis (Figure 1). We also investigated the hypothesis that the changes that were quantified in the retina before P21 Sema3 ligands in the SC can influence RGC axon guid- occurred during periods of RGC apoptosis, and synapse ance using an in vitro collapse assay on iso- generation and maturation. lated RGCs from a developmental stage when they would be growing into and within the SC in vivo. Changes in Class 3 Semaphorins and their receptor components during development of the rat superior Results colliculus Developmental expression profiles for the Sema3s and There were four qualitative groups of transcript expression their receptors are presented in Figures 2, 3, 4 and 5, levels in the SC: Plxna1 having the highest; followed by with those showing statistically significant changes in Sema3f, Nrp1,andNrp2;thenSema3b, Sema3c and Plxna3; expression explored further. Statistical comparisons il- and lower amounts for the remainder (Figures 4, and 5). lustrated in Figures 2, 3, 4 and 5 are summaries of sta- With the exception of L1cam, all of interest changed tistically significant peaks in expression; the entirety of expression significantlyduringSCmaturation. statistical comparisons can be found in Additional file 1: Sema3a RNA levels were highly variable at E16 (149% Table S1. Plxna3 in situ hybridisations at P1 and P7 are coefficient of variance; CV), peaked at E19, and a decline presented in Figure 6 showing co-expression in βIII- with increasing postnatal age. The expression profile of tubulin positive cells (presumptive RGCs). The effects of Sema3b differed from other transcripts; there was large exogenous recombinant Sema3s on RGC growth cone biological variation at many time points, with CVs greater collapse are presented in Figure 7, and data detailing ex- than 100%, and four apparent peaks in expression, only pression and detection of those recombinant Sema3s are one of which reached significance (E19). Sema3c tran- presented in Additional file 2: Figure S1. script expression peaked at E19 and P0, before falling to Sharma et al. BMC Developmental Biology 2014, 14:34 Page 3 of 14 http://www.biomedcentral.com/1471-213X/14/34

Figure 1 Developmental epochs in the rat neural retina and superior colliculus. Cytogenesis in the rat retina occurs from E9 through to P12, with distinct and overlapping waves of cell birth and differentiation [48–51]. Death of differentiated retinal ganglion cells (RGCs) is predominantly perinatal; starting at E20, peaking between birth and P3, falling sharply at P5 and continuing on until around P10 [52–57]. Astrocyte apoptosis is observed from P0, peaking between then and P5, and is associated with vasculature development [51,58]. Bipolar and amacrine cell apoptosis occurs concurrently with different peaks, from P5 until P13/P14 [54]. Synapse generation in the inner plexiform layer (IPL) starts at P2 with amacrine cells [59,60], and from P12 in bipolar cells [59], both peaking around the time of eye opening [54,59]. Outer plexiform layer (OPL) synaptogenesis begins at P4 [59–62] and, while its time-course in the rat is not certain, in the mouse continues until about P12 [63]. Neurogenesis in the rat superior colliculus (SC) occurs from E12 to E17 [64,65]. Apoptosis in the visual layers of the rat SC (stratum griseum superficiale; SGS) occurs from birth, peaking at between P4 and P7, and lasting until P13 [66,67]. Circuit maturation in the SGS, as judged by synapse density and composition, is mostly postnatal [68,69] in three stages: P0-P12/ P14 (eye opening), P12-P24 increasing number of synapses, P21-P30/P40 increasing ratio of inhibitory to excitatory terminals [70]. Retinal axons first enter the SC at E16/E17 [71], and continue to arrive until P5/P6 [52,72]. Cortico-tectal innervation starts at P3, with marked ingrowth between P6 and P12, and adult-like innervation observable from P18 [73]. Scale: downwards ticks mark 5 days, upwards ticks mark time points in this study; IPL: inner plexiform layer; OPL: outer plexiform layer; RGC: retinal ganglion cell; SGS: stratum griseum superficiale.

adult levels by around P14. In contrast to the other RNAs, all of which had significantly decreased RNA ex- Sema3s, Sema3e transcript levels increased steadily from pression by P0 and P7 that remained at low levels into E16 to P14, before a decrease at P21, then again increased adulthood. in the adult. Sema3f expression was highest at E16 and The patterns of expression of Sema3s in the developing E19, with RNA levels lower at birth and remaining at SC were substantially differentfromthoseseenintheretina, those reduced levels during postnatal development. There with peak expression occurring around or before birth. was relatively high biological variation in expression levels Temporally, these changes in transcript expression occurred of Nrp1, Nrp2 and all PlexinAs at E16 and E19 (CV > during the period of SC neurogenesis, synapse generation 100%) compared to other time-points. Nrp1 RNA expres- and maturation, and innervation of the SC by extrinsic affer- sion peaked at E19 then decreased steadily to P7 at levels ents including those from the retina and visual cortex. that were maintained into the adult. Nrp2 transcript levels peaked at E19 before falling steadily to E16 levels at P7 Expression of Plxna3 in the neonatal retina and beyond. Plxna1 RNA expression also peaked at E19, a Our qPCR data showed that peak Plxna3 RNA expres- pattern that was repeated for the remaining three PlexinA sion in the retina occurred during the period of peak Sharma et al. BMC Developmental Biology 2014, 14:34 Page 4 of 14 http://www.biomedcentral.com/1471-213X/14/34

Figure 2 RNA expression levels of the Class 3 Semaphorins in the rat retina during development. Sema3s: Sema3f is the most predominant transcript, and Sema3a the least. All transcripts increased expression through development and maintained relatively high levels into adulthood. Sema3a: Expression rises slightly to a peak at P14, dropping at P21 before rising again at adulthood. Sema3b: Expression is lowest in the embryonic retina, around half that found from P14 onwards. Sema3c: Transcription is stable up to P0 then rises twofold by P21 and into adulthood. Sema3e: Expression levels rise steadily after birth, reaching fourfold higher levels at their peak in the adult. Sema3f: There are two periods of increased expression, a doubling at birth before falling again by P7 and then rising back by P21 and into adulthood. Error bars are SEM; * p < 0.05, ** p < 0.01; n = 4–5.

RGC apoptosis (Figure 3). While rat retinal cells ex- the SC during this period (Figure 4). To further investi- pressing the Sema3 receptor components Nrp1, Nrp2, gate the capability of Sema3s to influence RGC axons as Plxna1 and Plxna2 have previously been identified using they grow into and within the SC we used a growth cone ISH [74] there has been no such characterisation for collapse assay to assess the capacity of developmentally Plxna3. We used ISH to examine the cellular location of appropriate RGCs to respond to Sema3 proteins. Re- Plxna3 transcripts in P1 and P7 rat retinas (Figure 6) and combinant Sema3 proteins (SEMA3A-GFP, SEMA3C- found that Plxna3 RNA was clearly expressed in the gan- FLAG, SEMA3E-FLAG, and SEMA3F-AP) in condi- glion cell layer at both these ages, with transcript expression tioned media were used to challenge immunopurified overlapping βIII-tubulin (RGCs) immunostaining. This is in P1/P2 RGCs in vitro. These proteins were detected in agreement with previously published data in the mouse ret- the conditioned media by western blot against their arti- ina [22]. Qualitatively, changes in the level of Plxna3 ex- ficial epitopes (Additional file 2: Figure S1A). Detected pression in our ISH data are consistent with the reduction bands for SEMA3A-GFP correspond to the approxi- in expression between P0 and P7 in our qPCR material. mately 130 kDa and 90 kDa bands reported previously [75]. SEMA3C-FLAG was detected at around 80 kDa as Growth cone collapse assay for purified neonatal RGCS expected, and also in a presumably processed form at in vitro around 70 kDa. Both SEMA3C-FLAG and SEMA3E- It has been suggested that Sema3s in the SC can assist FLAG were detected in media at approximately 100 kDa, in the guidance of ingrowing RGC axons [45], and our which might represent glycosylated forms of the proteins. data revealed increased expression of these ligands in SEMA3F-AP was detected at 150 kDa, in line with Sharma et al. BMC Developmental Biology 2014, 14:34 Page 5 of 14 http://www.biomedcentral.com/1471-213X/14/34

Figure 3 RNA expression levels of the Class 3 Semaphorin receptors in the rat retina during development. Receptors: Nrp1 transcript levels are around twofold higher than the peak level of Nrp2. Co-receptors: Plxna2 had the highest peak expression levels, and L1cam the lowest. Plxna2 and Plxna3 expression levels are very similar, with Plxna2 having around fourfold higher peak levels. Nrp2: RNA levels are lowest in the em- bryo, rising from P7 onwards and peaking at around fourfold higher in the adult. Plxna2: Expression peaks at birth at fivefold higher than at E16, drops again by P14 and is maintained at that level into the adult. Plxna3: RNA levels increase fourfold from E16 to their peak at birth before falling back to embryonic levels by P21. Plxna4a: Expression triples from E16 to P0, doubles again at P7, falls back one third by P21 before increasing again to peak in the adult. Error bars are SEM;* p < 0.05; n = 4–5. expectations. Transfected cells were analysed in culture by average percentage of the two replicates and statistical sig- epifluorescence and immunocytochemistry and were posi- nificance level presented in Figure 7A. Our data demon- tive for recombinant proteins while appropriate controls strate that neonatal rat RGCs are competent to respond to were negative (Additional file 2: Figure S1B-G). SEMA3A, SEMA3C, SEMA3E, and SEMA3F proteins at a Growth cones were visible as F-actin positive exten- developmental time point during which their axons are en- sions from the ends of βIII-tubulin positive neurites. tering into and growing within the SC in vivo [52,71,72]. Typically uncollapsed growth cones displayed the com- mon ‘hand like’ morphology, and collapsed growth cones Discussion appeared as thin stumps. Growth cones were counted We found statistically significant changes in ex- only at the end of neurites, even when significant growth pression of Sema3s and their co-receptors during matur- cone-like processes were observed along the shaft of a ation of the rat retina and SC. These changes differed in neurite, as these lateral extensions can occur in response degree and time-course between the retina and SC so to the collapse of the leading growth cone [76]. Exam- were not due to generalised or non-specific systemic ples of collapsed versus uncollapsed growth cones are il- changes. It is likely that the observed tissue-specific lustrated in Figure 7. Cultures treated with conditioned changes in expression reflect the biological roles of these media containing any of the Sema3s had significantly molecules and are in some way related to the development higher percentages of collapsed growth cones, with the and maturation of each structure [20,21,45]. While our Sharma et al. BMC Developmental Biology 2014, 14:34 Page 6 of 14 http://www.biomedcentral.com/1471-213X/14/34

Figure 4 RNA expression levels of the Class 3 Semaphorins in the rat superior colliculus during development. Sema3s: All Sema3 RNA expression levels, except for Sema3e, peak at E19 and are lowest by P21 and into the adult. Sema3b, Sema3c and Sema3f are the predominant transcripts with around 100 times higher levels than Sema3a and Sema3e. Sema3a: There is a high variability in expression at E16, peak expression at E19, falling twofold by birth and eightfold in the adult. Sema3b: Expression changes markedly throughout development, with high variability levels at both E19 and P14. Expression rises twenty-eightfold from E16 to E19 before falling back 17 fold by P7 and back to E16 levels at P21. Transcript levels then rise eightfold from P21 to adult. Sema3c: RNA levels peak at E19 and P0, falling twofold by P14 and maintaining that level into the adult. Sema3e: Transcript levels double and triple from the embryo to P7 and P14 respectively, dropping back at P21 and increasing again into the adult. Sema3f: RNA expression levels are highest in the embryo, dropping fivefold after birth and into the adult. Error bars are SEM; * p < 0.05; n = 4–5. data cannot localise these changes in expression to distinct differentiate and migrate within CNS tissues such as the cell populations, previous work in the retina suggests that cortex, cerebellum, and [39,77,78], and pre- the transcripts studied are heavily expressed in rat RGCs sumably they are able diffuse through tissues like other [27]. Importantly we also showed that neonatal rat RGCs guidance cues in the mammalian visual system such as are competent to respond to Sema3s at a developmental the Ephs and Netrin. Is the expression of Sema3s also in- time point when their axons are encountering the Sema3s volved in morphogenesis of the rat retina and SC? Retinal in the SC. We are wary of over-analysing temporal associ- neurogenesis occurs from E10/E11 through to the second ation data, but to help place our results in context we postnatal week [48], and while there is expression of all briefly discuss our data below with respect to known de- Sema3s and their receptors and co-receptors during much velopmental processes (Figure 1). of this period, expression levels are for the most part rela- tively low. Rod photoreceptor generation peaks at P0, as Class 3 Semaphorins and morphogenesis of the retina does transcript expression of Sema3f, Plxna2 and Plxna3 and superior colliculus RNAs. However there appears to be very little Sema3f Sema5s and Sema6s are known to mediate lamination of transcript expression in the outer nuclear layer (ONL) at the mouse retina, signalling through Plxna2, Plxna3 and this age [27], despite evidence of relatively high expression Plxna4 [22-24]. In addition, Sema3s guide cells as they of Sema3f RNA in the ONL of adult mice [79]. Sharma et al. BMC Developmental Biology 2014, 14:34 Page 7 of 14 http://www.biomedcentral.com/1471-213X/14/34

Figure 5 RNA expression levels of the Class 3 Semaphorin receptors in the rat superior colliculus during development. Receptors: Nrp1 and Nrp2 share similar expression profiles, with peaks around E19 and birth and then falling away into the adult. Co-receptors: All co-receptors bar L1cam follow the same expression profile, peaking at E19 before falling after birth and into the adult. Plxna1 and Plxna3 are the predominant transcripts. Nrp1: Peak RNA expression is at E19, falling threefold by P7 and maintained into the adult. Nrp2: Transcript levels double from E16 to P0, with high variability at E19. Expression falls from P0 to P7 and is maintained at around half peak levels in the adult. Plxna1: RNA levels are highly variable at E16 and E19 where they peak. From birth expression levels are halved, continue to decline up to P21 where they are sevenfold lower than at peak, and this level is maintained into the adult. Plxna2: RNA levels are highly variable at E16 and E19 where they peak. Expression drops threefold by P7 and is maintained into the adult. Plxna3: RNA levels are highly variable at E16 and E19 where they peak. Expression falls to half peak at P7, one quarter peak at P14 and is sevenfold lower than peak by P21 and in the adult. Plxna4a: RNA levels are highly variable at E16 and E19 where they peak before falling fourfold by P7 and into the adult. Error bars are SEM; * p < 0.05; n = 4–5.

On the other hand, rat SC neurogenesis occurs from early cytoarchitectural patterning of the SC. These data E12 to E17 [64,65] and the SC has an adult like laminar therefore suggest that the Sema3s might play a role in structure at P7 [80]. Sema3c and Nrp2 RNA expression patterning the SC, but not the retina. displayed the same upwards then downwards sweep as neurogenesis in the SC: elevated at E19 and P0, falling Class 3 Semaphorins and retinal ganglion cell apoptosis by P7. Similarly, Sema3a, Sema3b, and Sema3f as well as Sema3s, specifically Sema3a and Sema3f, have been Nrp1 and all PlexinA co-receptors showed peak expres- implicated in apoptosis of including RGCs sions at E19 that declined thereafter, coincidental with [81–86]. During retinal development, RGC apoptosis is Sharma et al. BMC Developmental Biology 2014, 14:34 Page 8 of 14 http://www.biomedcentral.com/1471-213X/14/34

Figure 6 Location of Plxna3 transcript in the neonate rat retina. Double in situ hybridisation (ISH) and immunofluorescence staining in P1 (A-C) and P7 retina (D-F). Expression was restricted mainly to the ganglion cell layer (GCL), with overlap of Plxna3 ISH and βIII-Tubulin positive putative retinal ganglion cells (C,F). GCL and dark arrows: ganglion cell layer; scale bars: 50 μm. predominantly perinatal [52–57] and we observed important for proper lamination of the inner plexiform three genes that had clear peaks in retinal RNA ex- layer (IPL) of the mouse retina, with mRNA expressed in pression at P0: Sema3f, Plxna2,andPlxna3.Interest- amacrine cells and heavily expressed in the IPL ingly, Plxna2 and Plxna3 are co-receptors for Sema3f [22–24]. We see no clear signal supporting this finding in and Sema3a [37], and Plxna3 has been shown to be the rat retina, although this may be a consequence of directly involved in Sema3a mediated cell death [84]. analysing global rather than cellular expression. In the Sema3a, Sema3f,andPlxna2 are expressed in neonatal SC, the majority of synaptogenesis is postnatal [68–70], rat RGCs [27], and we here show that in the rat Plxna3 and Sema3e RNA expression in the SC peaks during the RNA is predominantly expressed in the retinal gan- period when many synapses are being formed (P14) sug- glion cell layer (presumptive RGCs) at P1 (Figure 7). gesting a role for this molecule in SC synaptogenesis. ImportantlywealsoshowedthatP1/P2ratRGCswere responsive to both Sema3a and Sema3f (Figure 7). Our Effect of Sema3s on guiding RGC and other axons into data provide further suggestive evidence that the and within the superior colliculus Sema3s may influence RGC apoptosis. RGC axons enter the rat SC between E16 and P5/P6 [52,71,72], and previous results suggest that Sema3f may Class 3 Semaphorin expression during the period of contribute in helping to guide these axons to their cor- synapse generation and maturation in retina and superior rect retinotopic targets [45]. Furthermore, the Sema3 colliculus co-receptor L1cam is required for correct mapping of Sema3s are also involved in terminal axon branching, retinal axons in the mouse [93,94]. Previous data reveal synaptogenesis, pruning of terminal arbors, regulation of that during development of the mouse SC at least some synaptic plasticity, and maturation of [42,87– of the retino-tectal afferent fibres are positive for Nrp1, 92]. Synaptogenesis in the rat retina starts around P2 Nrp2,andL1cam [93–95], and P1 mouse RGC axons and is ongoing at P7 [54,59–62], when transcripts for express the receptors Nrp2 and Plxna1 [45]. While pre- Sema3c and Sema3e both become more abundant. Con- vious studies have shown responsiveness of some rodent versely, Sema3b and Sema3f RNAs increased after comple- RGCs to some Sema3s [45], these were not conducted at tion of retinal synaptogenesis (P14) [63]. The maintained a developmental time point where the RGC axons are relatively high transcript levels in the adult rat retina of growing into and within their Sema3 expressing targets Sema3b, Sema3c, Sema3e,andSema3f might indicate a [52,71,72]. Here we show for the first time that rat P1/ role in regulating synaptic plasticity in the mature retina. It P2 RGCs are capable of responding to exogenous Sema3 has been reported that Plxna2, Plxna3,andPlxna4 are proteins at an age when many of these neurons are Sharma et al. BMC Developmental Biology 2014, 14:34 Page 9 of 14 http://www.biomedcentral.com/1471-213X/14/34

Figure 7 Collapse assay of immunopurified P1/P2 rat retinal ganglion cells exposed to exogenous recombinant Sema3 proteins. P1 retinal ganglion cells (RGCs) were immunopurified and seeded in BD Culture Slides coated with PDL and laminin, grown overnight. The following day (effectively P2), RGCs were exposed to recombinant Sema3s for 30 minutes before fixation and immunostaining. Growth cones were considered collapsed if the F-actin did not extend past the end of the βIII-tubulin positive neurite with an arc of less approximately 60°. Only terminal growth cones were counted, because lateral extensions can occur in response to the collapse of the leading growth cone. Additionally, growth cones were only scored if their neurites were at least two widths long. All cultures treated with conditioned media containing recombinant Sema3 proteins showed statistically significantly higher numbers of collapsed growth cones when compared to cultures treated with control conditioned media (A). Also shown here are examples of neurons from each culture condition with growth cones scored as collapsed (hollow arrows) and uncollapsed (solid arrows). Solid arrows: uncollapsed growth cones; hollow arrows: collapsed growth cones; *** p < 0.001; scale bar: 20 μM. encountering these Sema3s as they grow axons into and suggestion that Sema3f may play a role in the formation within the SC. [52,71,72]. Our data show that almost all of the retinotopic map in the mammalian SC [45]. these Sema3s, bar Sema3e, had peak expression levels in Sema3s have also previously been shown to influence the the SC during this period of retinal afferent ingrowth, pattern of long-distance cortico-tectal projections [96]. Rat but that Sema3c and Sema3f had around tenfold higher visual cortical axons reach the SC at about P3, with marked expression compared to Sema3a and Sema3e. This dif- ingrowth between P6 and P12, and then final maturation ference in RNA expression levels might indicate that if between P12-P18 [73]. Our data show elevated Sema3e ex- these molecules are having an effect then the more likely pression during this period of ingrowth and maturation of candidates are Sema3c and Sema3f. However, some cau- cortical fibers in the SC, suggesting that Sema3e might in- tion must be used when extrapolating levels of RNA ex- fluence the development of corticotectal projections. pression to physiological significance; L1cam, Plxna3 and Plxna4a all had relatively low expression in the ret- Conclusions ina despite their known effects in patterning the retina In conclusion, the analysis of the Sema3s and their co- [22,23,93,94]. Nevertheless, our data do support the receptors in the developing rat visual system revealed a Sharma et al. BMC Developmental Biology 2014, 14:34 Page 10 of 14 http://www.biomedcentral.com/1471-213X/14/34

number of new and potentially important findings. In longer term storage. Four to five animals were used per the retina, mRNA expression levels changed for all group. Sema3s examined and there were age-specific changes in Nrp2, Plxna2, Plxna3,andPlxna4a mRNA expression. Tissue collection for in situ hybridisation In the SC there were also maturational changes in tran- P1 and P7 Wistar rats were anaesthetised by overdose script levels for all Sema3s, Neuropilins, and PlexinAs, injection of Lethabarb before transcardial perfusion with although the time-course of these changes differed 0.05% (w/v) heparin (David Bull Laboratories, Australia) markedly from those seen in the retina. These develop- in phosphate buffered saline (PBS), followed by 4% para- mental changes were associated with periods of RGC formaldehyde (w/v; Sigma-Aldrich, Australia) in 0.1 M apoptosis; neurogenesis in the SC; synapse generation, Sorenson’s Buffer (4% PFA; pH 7.4). Eyes and whole maturation, and plasticity in the retina and SC; and in- brains were dissected out and postfixed in 4% PFA for nervation by retino- and cortico-tectal axons in the SC. 30 minutes before cryoprotection in diethylpyrocarbo- Importantly, and consistent with a broad role for the nate (DEPC; Sigma-Aldrich, Australia) treated PBS con- Sema3 family in the maturing visual system, purified P1/ taining 30% sucrose (w/v; Sigma-Aldrich, Australia) P2 RGCs were sensitive to Sema3a, Sema3c, Sema3e, overnightat4°C.PBS/sucrosesolutionwasgradually and Sema3f mediated growth cone collapse in vitro. replaced with Jung tissue freezing medium (Leica These new data, providing a framework for future stud- Microsystems, Australia) over several days. After infusion ies aimed at elucidating such roles of the Sema3s. These tissues were snap frozen in isopropanol (2-propanol; new data describing the overall temporal regulation of Sigma-Aldrich, Australia). 20 μM sections were cut on a Sema3 expression in the rat retina and SC highlight mat- cryostat, placed on SuperFrost Plus slides (Menzel-Gläser, urational events that might be influenced by Sema3s, Germany), and stored at −80°C until processing by in situ providing a platform for further work characterising the hybridisation (ISH). functional impact of these proteins on mammalian visual system development. qPCR Methods Methodology for qPCR was as previously detailed [97], Experiments conformed to the Australian National Health and briefly described here. Total RNA was extracted and Medical Research Council (NHMRC) guidelines and with Tri Reagent (Molecular Research Center, USA) and were approved by the Animal Ethics Committee of The treated with recombinant DNase I (rDNase I; DNA-free; University of Western Australia. Animals were sourced Ambion, USA). First strand cDNA was synthesised using from the Animal Resources Centre (Western Australia). Omniscript (Qiagen, Australia) and random hexamers All statistical tests were performed using SPSS v19 (IBM, (Promega, Australia). Previously validated primer pairs USA). E0 was the day vaginal plug detected, P0 was date [97] were used to quantify RNA transcript expressions of birth. of Sema3a, Sema3b, Sema3c, Sema3e, Sema3f, Plxna1, Plxna2, Plxna3, Plxna4a, Nrp1, Nrp2, L1cam, and in- Tissue collection ternal reference genes Ppia, Rnr1, and Rpl19. qPCR runs Tissue collection for qPCR were performed on either a Rotor-Gene 3000 or 6000 Embryonic Wistar rats aged E16 or E19 were dissected (Qiagen, USA), using Bio-Rad iQ SYBR 2x Mastermix after caesarean section from mothers under halothane an- (Bio-Rad, Australia) in 10 μL reactions containing 500 aesthesia (5% (v/v) halothane (Rhone Merieux, Australia) nM of each primer, and Cq values obtained from using in 80:20 N2O:O2) and immediately placed in ice-cold F-10 the inbuilt second derivative maximum (SDM) equation. media (Gibco, Australia). Dams were sacrificed by intra- qPCR efficiencies are averages of efficiencies calculated peritoneal (IP) overdose (50 mg/100 g body weight) of from individual reactions using LinRegPCR [98,99]. Ini- pentobarbitone sodium (Lethabarb; Virbac, Australia). tial fluorescence levels (N0) were calculated from Cq and Postnatal and adult (8 to 10 week old) rats were anaesthe- mean efficiency values [100], and then normalised to the tised by overdose of Lethabarb (IP, 32.5 mg/100 g body geometric mean of appropriate internal reference genes weight). Eyes and SCs were quickly dissected out and [101]. BestKeeper analysis found all internal reference placed into fresh ice-cold F-10 media. Retinas were dis- genes were appropriate internal controls, as judged by sected free of the surrounding tunica and vitreous, left the standard deviation in Cq values. However, Rpl19 was and right retinas pooled and placed immediately into not used as an internal control because its RNA expres- RNAlater (Ambion, USA). Similarly, SCs were dissected sion changed significantly between developmental groups free of surrounding meninges and placed in RNAlater.All (Kruskal-Wallis ANOVA, p < 0.05 and p < 0.01 in the ret- tissues stored in RNAlater were cooled to 4°C within one ina and SC respectively), and was relatively poorly corre- hour for overnight storage, before transfer into −20°C for lated to the BestKeeper index SC (Spearman’sR=0.444). Sharma et al. BMC Developmental Biology 2014, 14:34 Page 11 of 14 http://www.biomedcentral.com/1471-213X/14/34

Thus expression data were normalised to the geometric Recombinant Class 3 Semaphorin conditioned media mean of Rnr1 and Ppia expression [101]. Expression plasmids (Sema3a-GFP, Sema3c-FLAG, Sema3e- Normalised data were analysed for changes between FLAG, Sema3f-AP) were also a gift from Professor Joost developmental groups by Kruskal-Wallis ANOVA (ana- Verhaagen. The Sema3a-GFP and Sema3f-AP plasmids lysis of variance). When Kruskal Wallis ANOVA re- have previously been characterised [75,102], containing sulted in p < 0.05 pairwise tests of statistical significance rat and mouse cDNA respectively. Sema3c-FLAG con- between time points were performed by Mann–Whitney tains mouse Sema3c cDNA, and Sema3e-FLAG con- U tests. tains human Sema3e cDNA. Plasmid constructs were verified by sequencing. HEK-293 T cultures were grown in DMEM: FBS In situ hybridisation (Dulbecco’s Modified Eagle Medium:fetal bovine serum; Plasmid for Plxna3 riboprobe was kindly provided by Gibco, Australia) 9:1 fortified with gentamycin (50 μg/mL; Professor Joost Verhaagen (Netherlands Institute for Invitrogen, Australia), and then grown without the pres- Neuroscience, Amsterdam). Sense and anti-sense probes ence of antibiotics overnight. Cells were transfected with were produced by restriction endonuclease (RE; New plasmids using Lipofectamine 2000 (Invitrogen, Australia), England Biolabs, USA) digest followed by in vitro RNA using Opti-MEM (Invitrogen, Australia) as per manufac- transcription (IVT) with digoxigenin labelled RNA (Roche, turer’s instructions. Media was changed to original growth Australia) and RNA polymerase (Roche, Australia). RE media with antibiotics 6 hours after transfection, and con- used were: sense, Kpn I; anti-sense, Spe I. Riboprobes were ditioned media collected 30 hours later. Control condi- hydrolysed for one hour at 60°C, precipitated with LiCl tioned medium was obtained by the same methods, using and 100% ethanol, resuspended in DEPC treated double sham (no plasmid) transfected cultures. deionised water (DDW), and stored at −80°C before use. P1 and P7 retinal sections were processed for ISH with Immunoprecipitation and detection of recombinant proteins the above riboprobe using methodology previously de- SEMA3A-GFP was purified from conditioned media using scribed [27], with α-dig-AP fragments (Roche, Australia) magnetic GFP-Trap beads (Chromotek, Germany). FLAG and NBT/BCIP (Roche, Australia). Sections were differen- tagged proteins (SEMA3C-FLAG and SEMA3E-FLAG), tiated in 70% ethanol if necessary, and stored in PBS before and Myc tagged proteins (SEMA3F-AP) were immuno- proceeding to immunohistochemistry (IHC). Sense ribo- precipitated from conditioned media using Dynabeads probes produced limited, diffuse staining, and all experi- Protein A/G (Invitrogen, Australia) and 5 μg α-FLAG mental runs showed positive anti-sense riboprobe staining antibody (Sigma-Aldrich, Australia), or 5 μg α-Myc anti- in control tissue (adult rat cerebellum; data not shown). body (CellSignaling, USA). Proteins were eluted in 2X Laemmili Buffer (250 mM Tris, 10% (v/v) glycerol, 4% (w/v) SDS, 2% (v/v) β-mercaptoethanol, 0.005% (w/v) Immunohistochemistry bromophenol blue, in DDW pH 6.8), separated by SDS- After ISH processing, sections were processed for im- PAGE (Mini-PROTEAN TGX Stain-Free Precast Gels; munofluorescence using standard methods. Antibody Bio-Rad, Australia) and transferred to nitrocellulose mem- diluent was 0.1% (w/v) Triton-X100, 10% (v/v) normal brane (Trans-Blot Turbo Mini Nitrocellulose Transfer goat serum (NGS; Chemicon, USA), in PBS. Primary Pack; Bio-Rad, Australia). Recombinant proteins were de- antibody (α-βIII-tubulin (TUJ1); mouse monoclonal, tected by western blot using 5% (w/v) skim milk pow- 1:2,000 dilution; Covance, USA) incubation was 4°C over- der in Tris buffered saline containing Tween (TBS-T; night in a humidified chamber, followed by secondary 100 mM Tris, 154 mM NaCl, 0.1% (v/v) Tween 20, in DDW antibody (α-mouse FITC, goat raised; 1:400 dilution; ICN pH 7.5) blocking buffer, α-GFP antibody (mouse monoclonal Cappel, USA) incubation for 2 hours in a dark humidified clones 7.1 and 13.1, 0.4 μg/mL;Roche,Australia),α-FLAG chamber at room temperature. Slides were coverslipped antibody (mouse monoclonal, 10 μg/mL; Sigma-Aldrich, using fluorescent mounting media (DAKO, Australia). Australia), α-Myc antibody (mouse monoclonal, 1:1,000 Cell cultures were stained for βIII-tubulin and F-actin. dilution; Cell Signaling, USA), and α-mouse-HRP sec- βIII-tubulin was labelled by IHC according to standard ondary antibody (1:10,000 dilution; Peirce Scientific, practice, using the same antibody diluent, incubation Australia). Proteins were visualised by chemilumines- times, and antibodies as described above. Following im- cence (Immun-Star; Bio-Rad, USA) using the Chemi- munostaining for βIII-tubulin, cultures were incubated Doc system (Bio-Rad, USA). with phalloidin-FITC (Sigma-Aldrich, Australia) diluted 1:1,000 in PBS for 2 hours at room temperature to stain Immunopurified P1/P2 retinal ganglion cell cultures F-actin, then were washed 3 times with PBS, before being P1 Wistar pups were euthanised by IP Lethabarb over- coverslipped using DAKO fluorescent mounting media. dose and retinas dissected free in PBS. Retinas were Sharma et al. BMC Developmental Biology 2014, 14:34 Page 12 of 14 http://www.biomedcentral.com/1471-213X/14/34

– dissociated with MiltenyiBiotec Tissue Dissociation Kit Additional file 2: Figure S1. Recombinant Sema3 protein expression in Postnatal Neurons (MiltenyBiotec, Australia), and RGCs conditioned media. A: Presented are three different blots, lined up against purified using the MACS RGC Isolation Kit (MiltenyiBio- the same ladder. Controls were from cultures that underwent a sham ’ transfection (no plasmid). Boxes indicated bands of expected molecular tec, Australia) according to manufacturer s instructions. weights. Detected bands for SEMA3A-GFP correspond to the approximately Purified RGCs were resuspended in growth media 130 kDa and 90 kDa bands reported previously [75]. SEMA3C-FLAG was (Neurobasal media (Invitrogen, Australia), 1X B27 supple- detected at around 80 kDa as expected, and also in a presumably μ processed form at around 70 kDa. Both SEMA3C-FLAG and SEMA3E- ment (Invitrogen, Australia), 1.25 M transferrin (Sigma- FLAG were detected in media at approximately 100 kDa, which might Aldrich, Australia), 0.2 μM progesterone (Sigma-Aldrich, represent glycosylated forms of the proteins. SEMA3F-AP was detected Australia), 100 μM putrescine (Sigma-Aldrich, Australia), at 150 kDa, in line with expectations. B-G: recombinant Class 3 Semaphorin μ expression in HEK-293 T cells transfected with expression plasmids for 230 nM sodium selenite (Sigma-Aldrich), 1.5 Mbovine Sema3a-GFP, Sema3c-FLAG, Sema3e-FLAG, and Sema3f-AP, as well as a no serum albumin (Sigma-Aldrich, Australia), 872 nM bovine plasmid control. SEMA3A-GFP expression was visualised directly by pancreas insulin (Sigma-Aldrich, Australia), 1 mM L- epifluorescence (B), SEMA3B-FLAG (C) and SEMA3E-FLAG (D) was μ detected following fluorescence immunocytochemistry, and SEMA3F- glutamine (Invitrogen, Australia), 30.6 MN-acetyl-cyst- AP was detected by the presence of colour reaction product from NBT/ eine (Sigma-Aldrich, Australia), 6 nM triiodithyronine (T3; BCIP (F). Scale bar: 50 μM. Sigma-Aldrich, Australia), 1 mM sodium pyruvate (Sigma- Aldrich, Australia)) fortified with growth factors (5 μM Competing interests insulin (Sigma-Aldrich, Australia), 5 μM forskolin (Sigma- The authors declare that they have no competing interests. Aldrich, Australia), 440 pM rat ciliary neurotrophic factor Authors’ contributions (CNTF; PeproTech, USA), 1.85 nM human brain-derived AS carried out qPCR, in situ hybridisations, immunohistochemistry, data neurotrophic factor (BDNF; PeproTech, USA) to increase analyses, project coordination, manuscript preparation and revision. CL survival as previously described [103]. Cells were seeded performed immunopreciptiation and western blotting. GWP participated in μ experimental design and manuscript revision. ARH conceived experimental onto PDL (100 g/mL; Sigma-Aldrich, USA) and mouse design, aided in project management, and contributed heavily to manuscript laminin (10 μg/mL; Invitrogen, Australia) coated BD Cul- revision. All authors read and approved the final manuscript. tureSlides (BD, Australia) at approximately 6,000 cells per 2 Acknowledgements mm . This study was supported by grants from the Western Australia Neurotrauma Research Program. Manuscript preparation was partly funded by a PhD Completion Scholarship from The University of Western Australia. We are Purified retinal ganglion cell growth cone collapse assay indebted to Professor Joost Verhaagen for his advice, and for the generous Purified RGC cultures were grown overnight at 37°C and gift of some of the reagents used in this study. 5% pCO . The following day (effectively P2) 30 μLofcondi- 2 Author details tioned media was added to the 300 μLofgrowthmediain 1School of Anatomy, Physiology and Human Biology, The University of 2 each well and incubated at 37°C and 5% pCO2 for 30 mi- Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. Stanford nutes. Cultures were then fixed with 4% PFA and stained Partnership for Spinal Cord Injury and Repair, Department of Neurosurgery, School of Medicine, Stanford University, Lorry I. Lokey Stem Cell Research for βIII-tubulin and F-actin. Epifluorescent photomicro- Building (SIM1), 265 Campus Drive, Stanford, CA 94305-5454, USA. graphs were taken of all neurons in all cultures, and encoded such that analysis was completed blind to treat- Received: 21 March 2014 Accepted: 15 July 2014 Published: 26 July 2014 ment group. Growth cones were scored as either collapsed or uncollapsed [104]: uncollapsed if the neurite itself was at References least two cell body widths in length, and the growth cone 1. Scicolone G, Ortalli AL, Carri NG: Key roles of Ephs and ephrins in retinotectal topographic map formation. Brain Res Bull 2009, 79:227–247. spread with an arc of at least approximately 60°; other 2. Erskine L, Herrera E: The retinal ganglion cell axon’s journey: insights into growth cones on neurites at least two cell body widths in molecular mechanisms of axon guidance. Dev Biol 2007, 308:1–14. length were classified as collapsed. The number of collapsed 3. Isenmann S, Kretz A, Cellerino A: Molecular determinants of retinal ganglion cell development, survival, and regeneration. Prog Retin Eye Res and uncollapsed growth cones was normalised to totals in 2003, 22:483–543. the control group, and compared using a Chi-square test. 4. Feldheim DA, O’Leary DDM: Visual map development: bidirectional The minimum number of growth cones counted per group signaling, bifunctional guidance molecules, and competition. Cold Spring Harb Perspect Biol 2010, 2:a001768. was 50, and the assay was repeated to confirm results. 5. McLaughlin T, Hindges R, O’Leary DD: Regulation of axial patterning of the retina and its topographic mapping in the brain. Curr Opin Neurobiol 2003, 13:57–69. Additional files 6. Reese BE: Development of the retina and optic pathway. Vis Res 2011, 51:613–632. 7. Harada T, Harada C, Parada LF: Molecular regulation of visual system Additional file 1: Table S1. Pairwise statistical comparisons of expression – data. Normalised data were analysed for changes between developmental development: more than meets the eye. Genes Dev 2007, 21:367 378. groups by Kruskal-Wallis ANOVA (analysis of variance). When Kruskal Wallis 8. Campbell DS, Regan AG, Lopez JS, Tannahill D, Harris WA, Holt CE: ANOVA resulted in p < 0.05 pairwise tests of statistical significance between Semaphorin 3A elicits stage-dependent collapse, turning, and branching in Xenopus retinal growth cones. J Neurosci 2001, 21:8538–8547. time points were performed by Mann–Whitney U tests. e: p = 0.05; * p < 0.05; ** p < 0.01. 9. Kita EM, Bertolesi GE, Hehr CL, Johnston J, McFarlane S: -1 biases polarization in the retina. Development 2013, 140:2933–2941. Sharma et al. BMC Developmental Biology 2014, 14:34 Page 13 of 14 http://www.biomedcentral.com/1471-213X/14/34

10. Halloran MC, Severance SM, Yee CS, Gemza DL, Raper JA, Kuwada JY: 34. Xiang R: Isolation of the human semaphorin III/F gene (SEMA3F) at Analysis of a Zebrafish semaphorin reveals potential functions in vivo. 3p21, a region deleted in lung cancer. Genomics 1996, Dev Dyn 1999, 214:13–25. 32:39. 11. Liu Y, Berndt J, Su F, Tawarayama H, Shoji W, Kuwada JY, Halloran MC: 35. Taniguchi M, Masuda T, Fukaya M, Kataoka H, Mishina M, Yaginuma H, Semaphorin3D guides retinal axons along the dorsoventral axis of the Watanabe M, Shimizu T: Identification and characterization of a novel tectum. J Neurosci 2004, 24:310–318. member of murine semaphorin family. Genes Cells 2005, 10:785–792. 12. Sakai JA, Halloran MC: Semaphorin 3d guides laterality of retinal ganglion 36. Stevens CB, Halloran MC: Developmental expression of sema3G, a novel cell projections in zebrafish. Development 2006, 133:1035–1044. zebrafish semaphorin. Gene Expr Patterns 2005, 5:647–653. 13. Callander DC, Lamont RE, Childs SJ, McFarlane S: Expression of multiple 37. Sharma A, Verhaagen J, Harvey AR: Receptor complexes for each of the class three semaphorins in the retina and along the path of zebrafish Class 3 Semaphorins. Front Cell Neurosci 2012, 6:28. retinal axons. Dev Dyn 2007, 236:2918–2924. 38. Kolodkin AL, Matthes DJ, O’Connor TP, Patel NH, Admon A, Bentley D, 14. Dell AL, Fried-Cassorla E, Xu H, Raper JA: cAMP-induced expression of Goodman CS: Fasciclin IV: sequence, expression, and function during neuropilin1 promotes retinal axon crossing in the zebrafish optic chiasm. growth cone guidance in the grasshopper embryo. 1992, 9:831–845. J Neurosci 2013, 33:11076–11088. 39. Roth L, Koncina E, Satkauskas S, Cremel G, Aunis D, Bagnard D: The many 15. Rosenzweig S, Raz-Prag D, Nitzan A, Galron R, Paz M, Jeserich G, Neufeld G, faces of semaphorins: from development to pathology. Cell Mol Life Sci Barzilai A, Solomon AS: Sema-3A indirectly disrupts the regeneration 2009, 66:649–666. process of goldfish optic nerve after controlled injury. Graefes Arch Clin 40. Yazdani U, Terman JR: The semaphorins. Genome Biol 2006, 7:211. Exp Ophthalmol 2010, 248:1423–1435. 41. Mizui M, Kumanogoh A, Kikutani H: Immune semaphorins: novel features 16. Luo Y, Shepherd I, Li J, Renzi MJ, Chang S, Raper JA: A family of molecules of neural guidance molecules. J Clin Immunol 2009, 29:1–11. related to collapsin in the embryonic chick nervous system. Neuron 1995, 42. Pasterkamp RJ, Giger RJ: Semaphorin function in neural plasticity and 14:1131. disease. Curr Opin Neurobiol 2009, 19:263–274. 17. Shepherd I, Luo Y, Raper JA, Chang S: The distribution of collapsin-1 mRNA 43. Suzuki K, Kumanogoh A, Kikutani H: Semaphorins and their receptors in in the developing chick nervous system. Dev Biol 1996, 173:185–199. immune cell interactions. Nat Immunol 2008, 9:17–23. 18. Takahashi T, Nakamura F, Jin Z, Kalb RG, Strittmatter SM: Semaphorins A 44. Neufeld G, Kessler O: The semaphorins: versatile regulators of tumour and E act as antagonists of neuropilin-1 and agonists of neuropilin-2 progression and tumour angiogenesis. Nat Rev Cancer 2008, 8:632–645. receptors. Nat Neurosci 1998, 1:487–493. 45. Claudepierre T, Koncina E, Pfrieger FW, Bagnard D, Aunis D, Reber M: 19. Henke-Fahle S, Beck KW, Püschel AW: Differential responsiveness to the Implication of neuropilin 2/semaphorin 3F in retinocollicular map chemorepellent Semaphorin 3A distinguishes Ipsi- and contralaterally formation. Dev Dyn 2008, 237:3394–3403. projecting axons in the chick midbrain. Dev Biol 2001, 237:381–397. 46. Kim J, Oh WJ, Gaiano N, Yoshida Y, Gu C: Semaphorin 3E--D1 signaling 20. Steinbach K, Volkmer H, Schlosshauer B: Semaphorin 3E/collapsin-5 regulates VEGF function in developmental angiogenesis via a feedback inhibits growing retinal axons. Exp Cell Res 2002, 279:52–61. mechanism. Genes Dev 2011, 25:1399–1411. 21. Steffensky M, Steinbach K, Schwarz U, Schlosshauer B: Differential impact of 47. Sefton AJ, Dreher B, Harvey A: Visual system. In The Rat Nervous System. 3rd semaphorin 3E and 3A on CNS axons. Int J Dev Neurosci 2006, 24:65–72. edition. Edited by Paxinos G. San Diego: Elsevier Academic Press; 22. Matsuoka RL, Chivatakarn O, Badea TC, Samuels IS, Cahill H, Katayama K, 2004:1083–1165. Kumar SR, Suto F, Chedotal A, Peachey NS, Nathans J, Yoshida Y, Giger RJ, 48. Rapaport DH, Wong LL, Wood ED, Yasumura D, LaVail MM: Timing and Kolodkin AL: Class 5 transmembrane semaphorins control selective topography of cell genesis in the rat retina. JCompNeurol2004, 474:304–324. Mammalian retinal lamination and function. Neuron 2011, 71:460–473. 49. Agathocleous M, Harris WA: Cell determination. In Retinal Development. 23. Matsuoka RL, Nguyen-Ba-Charvet KT, Parray A, Badea TC, Chedotal A, Edited by Sernagor E, Eglen S, Harris B, Wong R. Cambridge: Cambridge Kolodkin AL: Transmembrane semaphorin signalling controls laminar University Press; 2006:75–98. stratification in the mammalian retina. Nature 2011, 470:259–263. 50. Rapaport DH: Retinal Neurogenesis. In Retinal Development. Edited by 24. Sun LO, Jiang Z, Rivlin-Etzion M, Hand R, Brady CM, Matsuoka RL, Yau KW, Sernagor E, Eglen S, Harris B, Wong R. Cambridge: Cambridge University Feller MB, Kolodkin AL: On and off retinal circuit assembly by divergent Press; 2006:30–58. molecular mechanisms. Science 2013, 342:1241974. 51. Chan-Ling T, Chu Y, Baxter L, Weible Ii M, Hughes S: In vivo characterization 25. Kuwajima T, Yoshida Y, Takegahara N, Petros TJ, Kumanogoh A, Jessell TM, of astrocyte precursor cells (APCs) and astrocytes in developing rat retinae: Sakurai T, Mason C: Optic chiasm presentation of Semaphorin6D in the differentiation, proliferation, and apoptosis. Glia 2009, 57:39–53. context of Plexin-A1 and Nr-CAM promotes retinal axon midline crossing. 52. Dallimore EJ, Park KK, Pollett MA, Taylor JSH, Harvey AR: The life, death and Neuron 2012, 74:676–690. regenerative ability of immature and mature rat retinal ganglion cells 26. Semaphorin Nomenclature Committee: Unified Nomenclature for the are influenced by their birthdate. Exp Neurol 2010, 225:353–365. Semaphorins/Collapsins. Cell 1999, 97:551–552. 53. Dreher B, Potts RA, Bennett MR: Evidence that the early postnatal 27. de Winter F, Cui Q, Symons N, Verhaagen J, Harvey AR: Expression of class- reduction in the number of rat retinal ganglion cells is due to a wave of 3 semaphorins and their receptors in the neonatal and adult rat retina. ganglion cell death. Neurosci Lett 1983, 36:255–260. Invest Ophthalmol Vis Sci 2004, 45:4554–4562. 54. Horsburgh GM, Sefton AJ: Cellular degeneration and synaptogenesis in 28. Roche J, Boldog F, Robinson M, Robinson L, Varella-Garcia M, Swanton M, the developing retina of the rat. J Comp Neurol 1987, 263:553–566. Waggoner B, Fishel R, Franklin W, Gemmill R, Drabkin H: Distinct 3p21.3 55. Potts RA, Dreher B, Bennett MR: The loss of ganglion cells in the deletions in lung cancer and identification of a new human semaphorin. developing retina of the rat. Dev Brain Res 1982, 3:481–486. Oncogene 1996, 12:1289–1297. 56. Crespo D, O’Leary DD, Cowan WM: Changes in the numbers of optic 29. Kolodkin AL, Matthes DJ, Goodman CS: The semaphorin genes encode a nerve fibers during late prenatal and postnatal development in the family of transmembrane and secreted growth cone guidance albino rat. Brain Res 1985, 351:129–134. molecules. Cell 1993, 75:1389. 57. Perry VH, Henderson Z, Linden R: Postnatal changes in retinal ganglion 30. Luo Y, Raible D, Raper JA: Collapsin: a protein in brain that induces the cell and optic axon populations in the pigmented rat. J Comp Neurol collapse and paralysis of neuronal growth cones. Cell 1993, 75:217. 1983, 219:356–368. 31. Feiner L, Koppel AM, Kobayashi H, Raper JA: Secreted chick semaphorins 58. Fruttiger M: Development of the retinal vasculature. Angiogenesis 2007, bind recombinant neuropilin with similar affinities but bind different 10:77–88. subsets of neurons in situ. Neuron 1997, 19:539–545. 59. Dhingra NK, Ramamohan Y, Raju TR: Developmental expression of 32. Püschel AW, Adams RH, Betz H: Murine semaphorin D/collapsin is a synaptophysin, synapsin I and syntaxin in the rat retina. Brain Res Dev member of a diverse gene family and creates domains inhibitory for Brain Res 1997, 102:267–273. axonal extension. Neuron 1995, 14:941. 60. Sarthy PV, Bacon W: Developmental expression of a synaptic vesicle- 33. Sekido Y, Bader S, Latif F, Chen J-Y, Duh F-M, Wei M-H, Albanesi JP, Lee C-C, specific protein in the rat retina. Dev Biol 1985, 112:284–291. Lerman MI, Minna JD: Human semaphorins A (V) and IV reside in the 61. Kapfhammer JP, Christ F, Schwab ME: The expression of GAP-43 and 3p21.3 small cell lung cancer deletion region and demonstrate distinct synaptophysin in the developing rat retina. Brain Res Dev Brain Res 1994, expression patterns. Proc Natl Acad Sci U S A 1996, 93:4120–4125. 80:251–260. Sharma et al. BMC Developmental Biology 2014, 14:34 Page 14 of 14 http://www.biomedcentral.com/1471-213X/14/34

62. Weidman TA, Kuwabara T: Postnatal development of the rat retina. An 87. Schlomann U, Schwamborn JC, Muller M, Fassler R, Püschel AW: The electron microscopic study. Arch Ophthalmol 1968, 79:470–484. stimulation of dendrite growth by Sema3A requires integrin 63. Olney JW: An electron microscopic study of synapse formation, receptor engagement and focal adhesion kinase. J Cell Sci 2009, 122:2034–2042. outer segment development, and other aspects of developing mouse 88. Polleux F, Morrow T, Ghosh A: Semaphorin 3A is a chemoattractant for retina. Invest Ophthalmol 1968, 7:250–268. cortical apical dendrites. Nature 2000, 404:567–573. 64. Altman J, Bayer SA: Time of origin of neurons of the rat superior 89. Morita A, Yamashita N, Sasaki Y, Uchida Y, Nakajima O, Nakamura F, Yagi T, colliculus in relation to other components of the visual and visuomotor Taniguchi M, Usui H, Katoh-Semba R, Takei K, Goshima Y: Regulation of pathways. Exp Brain Res 1981, 42:424–434. dendritic branching and spine maturation by semaphorin3A-Fyn 65. Mustari MJ, Lund RD, Graubard K: Histogenesis of the superior colliculus signaling. J Neurosci 2006, 26:2971–2980. of the albino rat: a tritiated thymidine study. Brain Res 1979, 164:39–52. 90. Vo T, Carulli D, Ehlert EME, Kwok JCF, Dick G, Mecollari V, Moloney EB, 66. Giordano DL, Murray M, Cunningham TJ: Naturally occurring neuron death Neufeld G, de Winter F, Fawcett JW, Verhaagen J: The chemorepulsive in the optic layers of superior colliculus of the postnatal rat. J Neurocytol axon guidance protein semaphorin3A is a constituent of perineuronal 1980, 9:603–614. nets in the adult rodent brain. Mol Cell Neurosci 2013, 56:186–200. 67. Giordano DL, Cunningham TJ: Optic afferents, neuron maturation, and 91. Cioni JM, Telley L, Saywell V, Cadilhac C, Jourdan C, Huber AB, Huang JZ, neuron survival in the rat superior colliculus. Brain Res 1982, 256:365–368. Jahannault-Talignani C, Ango F: SEMA3A signaling controls layer-specific 68. Warton SS, Jones DG: Postnatal development of the superficial layers in branching in the cerebellum. Curr Biol 2013, 23:850–861. the rat superior colliculus: a study with Golgi-Cox and Kluver-Barrera 92. Uesaka N, Uchigashima M, Mikuni T, Nakazawa T, Nakao H, Hirai H, Aiba A, techniques. Exp Brain Res 1985, 58:490–502. Watanabe M, Kano M: Retrograde semaphorin signaling regulates 69. Warton SS, McCart R: Synaptogenesis in the stratum griseum superficiale synapse elimination in the developing mouse brain. Science 2014, of the rat superior colliculus. Synapse 1989, 3:136–148. 344:1020–1023. 70. Lund RD, Lund JS: Development of synaptic patterns in the superior 93. Demyanenko GP, Maness PF: The L1 cell adhesion molecule is essential colliculus of the rat. Brain Res 1972, 42:1–20. for topographic mapping of retinal axons. J Neurosci 2003, 23:530–538. 71. Bunt SM, Lund RD, Land PW: Prenatal development of the optic 94. Buhusi M, Schlatter MC, Demyanenko GP, Thresher R, Maness PF: L1 projection in albino and hooded rats. Brain Res 1983, 282:149–168. interaction with ankyrin regulates mediolateral topography in the – 72. Dallimore EJ, Cui Q, Beazley LD, Harvey AR: Postnatal innervation of the rat retinocollicular projection. J Neurosci 2008, 28:177 188. superior colliculus by axons of late-born retinal ganglion cells. Eur J 95. Gong S, Zheng C, Doughty ML, Losos K, Didkovsky N, Schambra UB, Nowak Neurosci 2002, 16:1295–1304. NJ, Joyner A, Leblanc G, Hatten ME, Heintz N: A atlas of 73. Thong IG, Dreher B: The development of the corticotectal pathway in the the central nervous system based on bacterial artificial . – albino rat: transient projections from the visual and motor cortices. Nature 2003, 425:917 925. Neurosci Lett 1987, 80:275–282. 96. Low LK, Liu XB, Faulkner RL, Coble J, Cheng HJ: Plexin signaling selectively 74. de Winter F, Oudega M, Lankhorst AJ, Hamers FP, Blits B, Ruitenberg MJ, regulates the stereotyped pruning of corticospinal axons from visual – Pasterkamp RJ, Gispen WH, Verhaagen J: Injury-induced class 3 cortex. Proc Natl Acad Sci U S A 2008, 105:8136 8141. semaphorin expression in the rat spinal cord. Exp Neurol 2002, 175:61–75. 97. Sharma A, Pollett MA, Plant GW, Harvey AR: Changes in mRNA expression 75. de Wit J, de Winter F, Klooster J, Verhaagen J: Semaphorin 3A displays a of class 3 semaphorins and their receptors in the adult rat retino- punctate distribution on the surface of neuronal cells and interacts with collicular system after unilateral optic nerve injury. Invest Ophthalmol Vis – proteoglycans in the extracellular matrix. Mol Cell Neurosci 2005, 29:40–55. Sci 2012, 53:8367 8377. 76. Davenport RW, Thies E, Cohen ML: Neuronal growth cone collapse 98. Ramakers C, Ruijter JM, Deprez RH, Moorman AF: Assumption-free analysis triggers lateral extensions along trailing axons. Nat Neurosci 1999, of quantitative real-time polymerase chain reaction (PCR) data. Neurosci – 2:254–259. Lett 2003, 339:62 66. 77. Tran TS, Kolodkin AL, Bharadwaj R: Semaphorin regulation of cellular 99. Ruijter JM, Ramakers C, Hoogaars WM, Karlen Y, Bakker O, van den Hoff MJ, morphology. Annu Rev Cell Dev Biol 2007, 23:263–292. Moorman AF: Amplification efficiency: linking baseline and bias in the analysis of quantitative PCR data. Nucleic Acids Res 2009, 37:e45. 78. Torre ER, Gutekunst C-A, Gross RE: Expression by midbrain dopamine 100. Schefe JH, Lehmann KE, Buschmann IR, Unger T, Funke-Kaiser H: Quantitative neurons of Sema3A and 3F receptors is associated with chemorepulsion real-time RT-PCR data analysis: current concepts and the novel “gene in vitro but a mild in vivo phenotype. Mol Cell Neurosci 2010, 44:135–153. expression’sCTdifference” formula. J Mol Med (Berl) 2006, 84:901–910. 79. Buehler A, Sitaras N, Favret S, Bucher F, Berger S, Pielen A, Joyal JS, Juan AM, 101. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Martin G, Schlunck G, Agostini HT, Klagsbrun M, Smith LE, Sapieha P, Stahl Speleman F: Accurate normalization of real-time quantitative RT-PCR data A: Semaphorin 3F forms an anti-angiogenic barrier in outer retina. FEBS by geometric averaging of multiple internal control genes. Genome Biol Lett 2013, 587:1650–1655. 2002, 3:Research0034. 80. Edwards MA, Caviness VS Jr, Schneider GE: Development of cell and fiber 102. Chen H, Chedotal A, He Z, Goodman CS, Tessier-Lavigne M: Neuropilin-2, a lamination in the mouse superior colliculus. J Comp Neurol 1986, novel member of the neuropilin family, is a high affinity receptor for the 248:395–409. semaphorins Sema E and Sema IV but not Sema III. Neuron 1997, 81. Shirvan A, Kimron M, Holdengreber V, Ziv I, Ben-Shaul Y, Melamed S, Melamed 19:547–559. E, Barzilai A, Solomon AS: Anti-semaphorin 3A antibodies rescue retinal 103. Meyer-Franke A, Kaplan MR, Pfieger FW, Barres BA: Characterization of the ganglion cells from cell death following optic nerve axotomy. JBiolChem signaling interactions that promote the survival and growth of 2002, 277:49799–49807. developing retinal ganglion cells in culture. Neuron 1995, 15:805–819. 82. Shirvan A, Ziv I, Fleminger G, Shina R, He Z, Brudo I, Melamed E, Barzilai A: 104. Kapfhammer JP, Xu H, Raper JA: The detection and quantification of Semaphorins as mediators of neuronal apoptosis. J Neurochem 1999, growth cone collapsing activities. Nat Protoc 2007, 2:2005–2011. 73:961–971. 83. Gagliardini V, Fankhauser C: Semaphorin III can induce death in sensory doi:10.1186/s12861-014-0034-9 neurons. Mol Cell Neurosci 1999, 14:301–316. Cite this article as: Sharma et al.: Changes in expression of Class 3 84. Ben-Zvi A, Manor O, Schachner M, Yaron A, Tessier-Lavigne M, Behar O: The Semaphorins and their receptors during development of the rat retina Semaphorin receptor PlexinA3 mediates neuronal apoptosis during and superior colliculus. BMC Developmental Biology 2014 14:34. dorsal root ganglia development. J Neurosci 2008, 28:12427–12432. 85. Bagnard D, Vaillant C, Khuth ST, Dufay N, Lohrum M, Püschel AW, Belin MF, Bolz J, Thomasset N: Semaphorin 3A-vascular endothelial growth factor- 165 balance mediates migration and apoptosis of neural progenitor cells by the recruitment of shared receptor. J Neurosci 2001, 21:3332–3341. 86. Bagnard D, Sainturet N, Meyronet D, Perraut M, Miehe M, Roussel G, Aunis D, Belin MF, Thomasset N: Differential MAP kinases activation during semaphorin3A-induced repulsion or apoptosis of neural progenitor cells. Mol Cell Neurosci 2004, 25:722–731.